JT effect in anti surge valve

[akbari1358]

Hi every body,
Is there any specific reference shows that joule-thomson phenomenon is considered/(or not considered) in design of compressor anti surge valve. The issue is related to natural gas trunk line compressor stations to boost the natural gas pressure by means of centrifugal turbo-compressors.
I asked this question in valve engineering forum but I didn't get a valuable response.

 

[LittleInch]

You did get a response which you haven't responded to.

There is no specific reference AFAIK.

However J T effect should be considered for this duty and calculation.

Cooling across a surge valve is not usually an issue as the gas is hot coming out of e compressor.

If you explain why you're asking the question and the background would help a lot.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[akbari1358]

Thanks, The background is that the outlet of the compressor is too hot and in surge case while opening ASV, the inlet temp will increase rapidly and will reach the compressor trip point due to high inlet temperature specially in full recycle case. So, if there is any consideration for temp drop in ASV then there is hope that compressor trip because of high inlet temp will not occur.

 

[georgeverghese]

The ASV feed should be from downstream of the compressor discharge cooler. So are you saying that the cooler is somehow thermally overloaded, hence resulting in high temperature recycle gas to the ASV? There can be many reasons for cooler thermal overload..

 

[akbari1358]

correct, the AC will be thermally overloaded because of recycle mode without considering temp drop in ASV.
So, do you agree that there is no consideration of temp drop in ASV by designer?

 

[LittleInch]

AC is air cooler?

Normally the ASV is sized for a flow to get you higher than surge not full flow. Either way your cooler should be sized for it but you should include for JT cooling.

Some numbers and data might help here.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[georgeverghese]

Think we've got a chicken or the egg situation here. At the moment, we cannot tell from the information you've provided so far, where the design flaw may be.

On the other hand, the designer should also consider the full recycle case when setting up the design thermal duty for the air cooler. This can become the controlling design case when the machine is operating fixed speed and the compressor suction temp is higher in full recycle mode than the temperature of the fresh feed gas. If this case is missed out, then you could have an air cooler overload situation, and higher compressor discharge temp also.

 

[akbari1358]

Okey, Let me explain you guys the exact position we may face with. The compressor unit is designed to deliver 15 MMSCMD to end users. The incoming gas temp is 18 degree Celsius and outlet temp after AC is 50 degree Celsius. Alarm incoming temp set point is 37 degree Celsius and HH temp will be 5 degree more. Based on seasonal transportation in summer case the delivery should be just 2MMSCMD (near full recycle mode) without any change in delivery pressure and this is the scenario I cant analyse
inlet/outlet pressure : 54/120 bar
comp. in/out size: 24"/24"
recycle line size: 18"

 

[LittleInch]

So how are you controlling flow? A 7:1 turndown is a lot for one machine.

If you're trying to do this by opening up the A/S valve to 100% that's a big waste of energy and doesn't sound like a good plan to me.

can you past the compressor curve and some air cooler data?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[georgeverghese]

The description and the recycle line size seems to indicate a fixed speed compressor. This operating mode may lead to a higher compressor discharge temp, and with a recycle flow of approx 13mmscmd may be the design thermal case for the air cooler in summer ??

Simulating this recycle case to reach convergence shouldnt be too difficult. You could set up the air cooler in rating mode on the simulator to see if it is up to this duty in the one simulation.

 

[akbari1358]

unfortunately it's not possible to share project doc on an internet forum, however the main question was about joule-thomson phenomenon in ASV.

 

[LittleInch]

A centrifugal machine, even foxed speed should allow flow to be controlled by a discharge control valve until you hit the surge line. A 15 mscmd machine should be able to be throttled to about 6 or 7 mmscmd workout having to use the ASV.

You can always hand sketch the system or black out project info on a compressor curve before scanning it.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.